Bit manipulation (set and clear) descriptions

Created by: DavidM-EMUS

In the documentation, updated descriptions of the functionality of p.bclr, p.bclrr, p.bset, and p.bsetr are unclear and possibly incorrect. The documentation lists the functionality of p.bclr as: "p.bclr | rD, rs1, Is3, Is2 | rD[min(Is3+Is2,31):Is2] = 0 the rest of the bits of rD are passed through and are not modified" at https://core-v-docs-verif-strat.readthedocs.io/projects/cv32e40p_um/en/latest/instruction_set_extensions.html#bit-manipulation-operations This description is confusing and/or misleading for three reasons: 1; If the instruction performs as specified above, what is the purpose of the rs1 argument? 2: The instruction didn't perform in the manner specified above in the past. 3: The instruction doesn't perform as specified in testing. The instruction presently copies rs1 to rD and then performs as specified in the documentation. For example, under the present description, p.bclr 0x82f98620, 0x12577823, 8, 24 would output 0x00f98620, when testing shows that the result is 0x00577823.

These issues are also the same for p.bclrr, p.bset, p.bsetr.

Recommended fix: Add to description to denote that rD is set to the value of rs1 with the specified bits cleared. Example p.bclr: rD = (rs1 & ~(((1<<ls3)-1)<<ls2)) Example for p.bclrr: rD = (rs1 & ~(((1<<rs2[9:5])-1)<<rs2[4:0])) Example for p.bset: rD = (rs1 | (((1<<ls3)-1)<<ls2)) Example for p.bsetr: rD = (rs1 | (((1<<rs2[9:5])-1)<<rs2[4:0]))